Remove legacy Ollama preprocessing pipeline#70
Merged
Conversation
Remove the optional Ollama local LLM preprocessing pipeline (3 agents: ingestion, processing, synthesis) and all related infrastructure. The MCP host (Claude Code / Claude Desktop) handles aggregation directly via the aggregate_results tool, making the Ollama pipeline redundant. Deleted: - Ollama MCP server, Dockerfiles, entrypoint scripts - Agent servers (ingestion, processing, synthesis) - Agent prompt templates - Ollama-specific tests Updated: - docker-compose.yml: removed 5 Ollama services and volume - Makefile: removed Ollama targets and profile references - Config: removed ollama_url/ollama_model settings - Models: renamed ollama_model to model in AggregatedMetadata/graph - Reports: updated generators to use generic "Model" label - CI: removed Ollama/agent matrix entries from build workflow - Docs: cleaned README.md, DOCKER.md, CLAUDE.md - Dependencies: removed ollama, fastapi, uvicorn from requirements All 188 tests pass. https://claude.ai/code/session_01BaZKkxAotm6KAehzs2qYjL
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
This PR removes the entire legacy Ollama-based preprocessing pipeline, including the MCP server orchestrator and three agent containers (Ingestion, Processing, Synthesis). The MCP host (Claude) now handles data aggregation directly via the
aggregate_resultstool, which is faster and more accurate than the previous multi-container approach.Key Changes
mcp_servers/ollama_mcp_server.py— the thin HTTP orchestrator that called three agent containers sequentiallyblhackbox/agents/base_agent_server.py— FastAPI base server for agentsblhackbox/agents/base_agent.py— base class for Ollama agentsblhackbox/agents/ingestion_agent.py,processing_agent.py,synthesis_agent.py— individual agent implementationsblhackbox/agents/ingestion_server.py,processing_server.py,synthesis_server.py— agent container entry pointsblhackbox/prompts/agents/ingestionagent.md,processingagent.md,synthesisagent.mddocker/ollama-mcp.Dockerfile,docker/ollama.Dockerfiledocker/agent-ingestion.Dockerfile,docker/agent-processing.Dockerfile,docker/agent-synthesis.Dockerfiledocker/ollama-entrypoint.shtests/test_ollama_mcp.py,tests/test_agent_server.py,tests/test_agents.pyollama-mcp,agent-ingestion,agent-processing,agent-synthesis, andollamaservices; removed--profile ollamadocumentation.env.exampleollama_urland related settings fromblhackbox/config.pyrequirements.txt(ollama, fastapi, uvicorn, pydantic)README.md,DOCKER.md,CLAUDE.md.github/workflows/build-and-push.ymlblhackbox-mcp.jsonandblhackbox-mcp-catalog.yamlImplementation Details
AggregatedPayloadusing theaggregate_resultstoolhttps://claude.ai/code/session_01BaZKkxAotm6KAehzs2qYjL